Re: Is supertyping orthadox?

Nicholas Geovanis <nickgeo_at_merle.acns.nwu.edu> wrote:
: If I had to bet on it, I'd suspect that the consultants are going to be
: using a homegrown set of query-building and/or screen-building tools
: (which they haven't told you about yet :-). The tools will be issuing
: queries based on their (already coded-for) "meta-schema" in order to do
: their work, which of course will be the work of generating code to
: implement the screens and perhaps reports, perhaps even a generic querying
: layer between the application proper and the DBMS if the toolset is
: ambitious.

And of course they will be billing you for "time spent writing query tools". I don't envy the DBAs trying to provide good performance for these 'generic' queries (ie, queries written by tools written for the consultants' previous DBs..).

Aside from that, I wonder about the solution that stared this thread. I wrote a meta-schema with some supertyping facilities, but not as much, nor as rigidly, as the consultants. The major complaints sound like an inability to define constraints, triggers, & relations. Are these the result of breaking the relational model, or that the consulant's tools are weekend hacks (and lack mature options)?

Can anyone answer that, and say how *they* would write their ideal metaschema? Received on Thu Mar 22 2001 - 00:29:32 CET